ABSTRACT Targeted molecular agents are a major advance in cancer therapy. The tyrosine kinase inhibitor imatinib mesylate targets the mutant KIT protein in gastrointestinal stromal tumor (GIST). However, while imatinib is highly effective against metastatic GIST, it almost never induces a complete response. Tumor progression occurs at a median of approximately 20 months and most patients eventually die from their disease. We have found that a novel tyrosine kinase inhibitor called PLX3397 is substantially more effective than imatinib in a genetically engineered mouse model of GIST and in human xenografts. After treatment, there is just a small fraction of residual tumor cells. Given that we have previously found that the immune system contributes to the efficacy of imatinib in GIST, we propose in Aim 1 to define the immune response after PLX3397 therapy in mouse GIST. Since PLX3397 causes much greater tumor destruction than imatinib, we hypothesize that the anti-tumoral immune response is greater. To eradicate the few remaining cells after PLX3397 treatment, we will combine it with the immunomodulatory agents anti-CTLA-4 or anti-PD-1 in Aim 2. The effect on tumor size and the antigen-specific CD8 T cell response will be assessed. In Aim 3, we will determine whether human intratumoral CD8 T cells that are freshly isolated from surgical specimens are functional against GIST. We anticipate that our findings will increase our understanding of GIST and may establish the rationale for a novel clinical trial combining molecular and immune therapy in GIST. We expect that our work will not only improve the lives of patients with these tumors, but also be relevant to other cancers in which targeted molecular therapy is used.